Dear Tomonori, > Hmm, why my Signed-off-by is here? > > Anyway, I'll wait for an ACK from one of the guys listed above. I'm very sorry but it seems I made a mistake when copy-and-paste addresses from other commit log. I'll fix the commit log then resend the patch as v2. Best regards, Takashi > -----Original Message----- > From: FUJITA Tomonori [mailto:fujita.tomonori@xxxxxxxxxxxxx] > Sent: Thursday, October 13, 2016 6:15 PM > To: menjo.takashi@xxxxxxxxxxxxx > Cc: stgt@xxxxxxxxxxxxxxx; ishizaki.teruaki@xxxxxxxxxxxxx; > mitake.hitoshi@xxxxxxxxxxxxx > Subject: Re: [PATCH] sheepdog: reject snapshot as a LUN to avoid deleting > working VDI > > On Thu, 13 Oct 2016 17:49:04 +0900 > Takashi Menjo <menjo.takashi@xxxxxxxxxxxxx> wrote: > > > This commit lets the sheepdog driver reject forms of filename below. > > This is to avoid unintended deletion of working VDI when a snapshot is > > given as a LUN. > > > > * unix:<path_of_unix_domain_socket>:<vdi>:<tag> > > * unix:<path_of_unix_domain_socket>:<vdi>:<snapid> > > * tcp:<host>:<port>:<vdi>:<tag> > > * tcp:<host>:<port>:<vdi>:<snapid> > > > > In the case that a snapshot is given as a LUN by any of them above, the > > sheepdog driver attempts to delete it then create a new writable VDI when > > some data is being written to it. The deletion request contains name and > > VDI ID of the snapshot. However, sheep daemon uses its name but ignores > > VDI ID and, what is worse, the request contains neither snapshot ID nor > > tag. So sheep daemon receiving the request interprets it as deletion of > > working VDI, not snapshot. If there is working VDI with the same name, > > it will be deleted unintentionally. > > > > I decided not to fix the sheepdog driver to send snapshot ID or tag but > > to let it reject snapshot as a LUN because, even if I fix it, creating > > a new writable VDI will be failed when working VDI exists. If you want > > workaround, use "dog vdi clone" command to create a new writable VDI from > > snapshot then give the new VDI as a LUN. > > > > Cc: Teruaki Ishizaki <ishizaki.teruaki@xxxxxxxxxxxxx> > > Cc: Hitoshi Mitake <mitake.hitoshi@xxxxxxxxxxxxx> > > Signed-off-by: Takashi Menjo <menjo.takashi@xxxxxxxxxxxxx> > > Signed-off-by: FUJITA Tomonori <fujita.tomonori@xxxxxxxxxxxxx> > > Hmm, why my Signed-off-by is here? > > Anyway, I'll wait for an ACK from one of the guys listed above. > -- To unsubscribe from this list: send the line "unsubscribe stgt" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html